AWSTemplateFormatVersion: 2010-09-09
Description: >-
  This template provisions a AWS Glue crawler and AWS Glue database to help extract external table defintions 
  required for using Amazon S3 as a source in AWS DMS.  
Parameters:
  SourceS3Bucket:
    Description: Name of the Amazon S3 Bucket where the source files reside
    Type: String
  SourceS3FolderPath:
    Description: >-
      Amazon S3 prefix where source files reside.  The files in this folder should be in the form <schema name>/<table name>.  
      Refer to https://docs.aws.amazon.com/dms/latest/userguide/CHAP_Source.S3.html for more information
    Type: String
Resources:
  DMSS3AsSourceTableDefGlueRole:
    Type: 'AWS::IAM::Role'
    Properties:
      AssumeRolePolicyDocument:
        Version: 2012-10-17
        Statement:
          - Effect: Allow
            Principal:
              Service:
                - glue.amazonaws.com
            Action:
              - 'sts:AssumeRole'
      Path: /
      Policies:
        - PolicyName: DMSS3AsSourceTableDefGluePolicy
          PolicyDocument:
            Version: 2012-10-17
            Statement:
              - Action:
                  - 'Glue:*'
                Resource:
                  - '*'
                Effect: Allow
              - Effect: Allow
                Action:
                  - 'logs:CreateLogGroup'
                  - 'logs:CreateLogStream'
                  - 'logs:PutLogEvents'
                Resource:
                  - 'arn:aws:logs:*:*:/aws-glue/*'
              - Effect: Allow
                Action:
                  - 's3:ListBucket'
                Resource:
                  - !Join 
                    - ''
                    - - 'arn:aws:s3:::'
                      - !Ref SourceS3Bucket
              - Effect: Allow
                Action:
                  - 's3:GetObject'
                Resource:
                  - !Join 
                    - ''
                    - - 'arn:aws:s3:::'
                      - !Ref SourceS3Bucket
                      - /
                      - !Ref SourceS3FolderPath
                      - /*
  DMSS3AsSourceTableDefDatabase:
    Type: 'AWS::Glue::Database'
    Properties:
      CatalogId: !Ref 'AWS::AccountId'
      DatabaseInput:
        Name: !Join 
          - ''
          - - dms_db_
            - !Ref 'AWS::StackName'
            - _
            - !Ref SourceS3FolderPath
        Description: Database contains all tables/files to be loaded by DMS
  DMSS3AsSourceTableDefCrawler:
    Type: 'AWS::Glue::Crawler'
    Properties:
      Name: !Join 
        - ''
        - - dms_crawler_
          - !Ref 'AWS::StackName'
      Role: !GetAtt 
        - DMSS3AsSourceTableDefGlueRole
        - Arn
      DatabaseName: !Ref DMSS3AsSourceTableDefDatabase
      Targets:
        S3Targets:
          - Path: !Join 
              - ''
              - - !Ref SourceS3Bucket
                - /
                - !Ref SourceS3FolderPath
      SchemaChangePolicy:
        UpdateBehavior: UPDATE_IN_DATABASE
        DeleteBehavior: LOG
  
Outputs:
  DMSDatabase:
    Description: 'Name of the Glue Database where the schema/tables have been identified '
    Value: !Ref DMSS3AsSourceTableDefDatabase
  DMSGlueCrawler:
    Description: 'Name of the Glue crawler'
    Value: !Ref DMSS3AsSourceTableDefCrawler